Research Engineer - MOIRA - Automatic Multi-sensor Validation Methods

Siemensglobal (Leuven VBR, België) 6 dagen geleden gepost



















Marie Curie Early Stage Researcher – MOIRA – Automatic multi-sensor
validation methods





Introduction:



Siemens Digital Industries Software (SISW),
headquartered in Leuven, Belgium, will have an open PhD position in frame of
the European Training Network on Monitoring Large-Scale Complex Systems (“MOIRA”),
funded by the European Commission
through the H2020 “Marie Skłodowska-Curie Innovative Training Networks” (ITN)
program.



The objective of the MOIRA project is to develop the
next generation of knowledge discovery methodologies, algorithms and
technologies, so enabling data-driven, plant-wide fleet monitoring, with the focus
on real-time diagnostics and prognostics. This objective will be achieved by
having 15 early-stage researchers (ESRs) in a collaborative network between top
European universities, research institutes, wind-turbine and plant operators,
OEMs and industrial stakeholders with an expertise in mechanical engineering,
computer science, signal processing, vibrations, inverse problems, operations
maintenance, data analytics and networks.



The PhD project connected to this vacancy involves
SISW as lead beneficiary and the KU Leuven (KUL) as the academic partner and PhD
awarding entity. The ESR will become part of the research team of the SISW TEST
division and will cooperate closely with the SISW staff as well as other
international visiting researchers and students. The ESR will be enrolled as a
PhD student in the KUL doctoral school.





PhD Project Description:



The ESR will research methods that enable the
automatic detection of “incorrect” sensor data. Sensors are exposed to tough
operating conditions in many industrial environments (e.g., excavation machines
driving on off-road tracks, gantry cranes in steel mills, etc.). Therefore, a
common problem is the occurrence of “measurement anomalies”, i.e., where part
of the data is incorrect in the sense that there are some deviations from what
was intended to be measured. Examples of measurement anomalies with particular
shapes are dropouts, offsets, drifts and spikes, but the measurement anomaly
can also be a more subtle problem with the data. An advanced automatic sensor
validation method is thus highly sought after.



The ESR will investigate machine-learning methods that
are trained to recognize incorrect sensor data. A systematic approach will be followed:
in the first stage, a supervised learning approach will be adopted, whereby it
is assumed that an historical dataset with fully labelled examples is
available. As this assumption might not prove to be practically realizable in many
cases, an unsupervised anomaly-detection approach will be investigated in the
second stage. Such an approach does not require labelled data, but is typically
more difficult to implement successfully compared to a supervised approach. An
interesting third alternative that will be investigated is a semi-supervised
approach, where a small labelled dataset (e.g., obtained from expert user
feedback) is available in addition to the larger unlabelled dataset. Besides the
systematic investigation outlined above (supervised – unsupervised –
semi-supervised), a particular focus point will be to leverage the fact that
there are multiple sensors, i.e., there is a certain redundancy in the
measurement setup so that some sensors will be measuring related quantities.
While measurement anomalies are non-physical events that occur at random times
(so that they will likely not be observed in multiple sensor channels), real
physical events likely affect multiple (closely located) sensors. A comparison
between sensor pairs (e.g., linear or nonlinear correlation analysis) could
thus be exploited in order to better detect the measurement anomalies (for
example, to distinguish an incorrect measurement spike from a true physical
shock event in the data).





Timeline and remuneration: The earliest start time is 1st
March 2021. The Marie Curie grant foresees funding for a duration of 3 years,
however, given that a typical PhD duration in Belgium is 4 years, extra funds
are reserved such that a fourth PhD year can be added. Furthermore, in order to
stimulate intersectoral and international mobility, the ESR will have short
research visits (so-called “secondments”) to at least two Beneficiary/Partner
Organisations (with the secondments not exceeding 30% of the duration of the doctoral
training).



The remuneration is generous and will be in line with
the EC rules for Marie Curie grant holders. It consists of a salary augmented
by a mobility allowance, resulting in a net monthly salary of about 1900-2300
Euro depending on family status.





Supervisors
and main contacts:



Siemens Digital Industries
Software: dr. Bram Cornelis (research manager)



KUL: prof. Konstantinos
Gryllias, prof. Wim Desmet



Candidate
Profiles:



Applicants
must have a MSc degree or equivalent in mechanical/mechatronic engineering or related field, which will
qualify for starting a PhD programme.



They
must have:



·      
Excellent
qualification in engineering disciplines such as mechanics, electronics,
physics and mathematics;



·      
Very
strong interest in machine learning



·      
Experience
with scientific computing and high-level programming languages such as Matlab
or Python.



·      
Affinity
with the scientific research methodology;



·      
Interest
to develop and implement a long-term research programme leading to a PhD;



·      
Capability
to work independently and in a team;



·      
Fluent
in spoken and written English;





Competences that are
considered as an additional advantage:



·       Previous hands-on
experience with machine learning (incl. deep learning) is a huge asset.



·      
Solid
background in experimental vibration and/or acoustic testing and signal
processing is an advantage.



·      
Previous
hands-on experience with large channel count measurement campaigns, such as in
automotive, heavy industries or aerospace testing facilities will be an
advantage.





Marie Curie eligibility
Criteria


To be eligible, you need to be an
"early stage researcher" i.e. simultaneously fulfill the following
criteria at the time of recruitment:





·       Mobility: you must not have resided or carried out your main
activity (work, studies, etc...) in Belgium for more than 12 months in the 3
years immediately prior to your recruitment under the MOIRA project.



·       Qualifications and research experience: you must be in the
first 4 years of your research career after the master degree was awarded.











Organization: Digital Industries

Company: Siemens Industry Software NV

Experience Level: Experienced Professional

Job Type: Full-time

Research Engineer - MOIRA - Automatic Multi-sensor Validation Methods

Apply On Company Site
Back to search page
;